On 5/14/2010 4:38 PM, Linda Walsh wrote:
Marc Chamberlin wrote:
I have a problem that I occasionally see on all my systems,
...
Has anyone else seen this problem and is there a solution to it? I searched bugzilla but could not find any complaints about this issue, and since I am seeing it on several different computers, I find it hard to believe that I am the only one seeing it.. I am running openSuSE 11.2, x86_64 on all my systems. I also looked in /var/log/messages but see nothing that indicates a problem.
---- I have a similar problem but didn't report it for 2 reasons: 1) I can't figure out where the problem is, and 2) (as a result of 1) can't figure out if it is related to some configuration problem I've created.
But in my case the outcome is worse -- the system hangs when shutting down. It kills off all processes and goes into the final stages of shutdown -- but hangs before it actually unmounts the disks. It has to be power-cycled -- the result is disk-thrashing. I've lost a software RAID5 array twice, this way (that contained backups). Neither of my hardware RAID5's have been affected.
It does indicate to me that Software RAID5 isn't very reliable and shouldn't be used for important data. I have experienced some data loss when some local reconfigurations caused loss of primary data when backups had been on a software RAID5.
I'm contemplating how to switch over to hardware RAID5 for my backups so they won't be destroyed so often.
But had my systems not experienced the shutdown-reboot problems you mentioned, I might never have noticed a problem with software RAID.
FWIW -- if files didn't get written out completely, that'd be 'fine', but to have the arrays fail to be able to be assembled even in degraded mode is unacceptable from a reliability standpoint.
-linda
FWIW I have seen another aspect of what I consider is a serious design flaw of the Linux kernel, in the shutdown/reboot processes. Some programs (MythTV comes immediately to my mind) can actually intercept the shutdown/reboot request and prevent it! And man does that infuriate me when I see it happen!!! When the OS receives a shutdown or reboot request, NOTHING short of a serious hardware failure should prevent it, IMHO. And NO app or process should EVER be allowed to prevent it. A shutdown or reboot request may have been initiated in response to an emergency, (especially in a situation like mine where I am remotely administering a computer and cannot be there to hit the power down button) therefore the Linux kernel must absolutely honor the request and force the shutdown or reboot to take place, regardless of any protests/failure from some darn app or process! I kinda suspect there has been some creeping features that are now allowing apps to intercept the shutdown/reboot request, and like in my situation, stopping the process at a runlevel 3 terminal with a login prompt. This is DEAD WRONG, it strongly indicates a fundamental design flaw in the kernel, because it should never ever happen. Best efforts should be made to preserve data, but that does not override the need to expediently shut down or restart a computer when so requested. In the scenario Linda points out, I too have seen a failure occur when unmounting devices, resulting in a hung system. Again that is just dead wrong, and the Linux kernel should prevent such an event by timing out the hung process and forcing the system to proceed with the shutdown/reboot process. Marc...